Colorectal cancer (CRC) is an extremely prevalent and difficult to treat cancer. In the advanced stage of disease there are virtually no effective treatment options rendering it the second-leading cause of cancer- related deaths in the developed world. Unlike a number of other cancer treatment strategies, which have benefited from the development of targeted agents, the standard of care for CRC remains a combination of cytotoxic chemotherapies. There is an urgent need for more targeted, less toxic, treatments. Defining the tumor-specific requirements for cancer growth and progression offers one potential avenue for drug target discovery. This proposal integrates new genetic technologies and ex vivo cultures systems to characterize the genetic requirements for CRC maintenance and progression, with the goal of identifying tumor-specific weaknesses that could be exploited for cancer therapy. Mutational inactivation of the Adenomatous Polyposis Coli (APC) tumor suppressor is thought to be the initiating event in most familial and sporadic colon cancers. APC disruption results in uncontrolled activation of the Wnt pathway and because of this there is currently significant effort to develop selective Wnt signalling inhibitors for treatment of CRC. However, progression from benign adenoma to CRC involves many additional genetic and epigenetic alterations, most commonly, loss of the p53 tumor suppressor (60%) and/or oncogenic activation of Kras (45%). Given the complex nature of the disease it is unclear if CRCs remain dependent on APC loss and Wnt hyperactivation for growth and survival, and thus whether Wnt- targeted therapies will be effective in treating advanced malignancies. Here I will address this outstanding, but important question, using a unique transgenic mouse model I developed that allows temporal, spatial and reversible control of APC expression. The approach described combines existing and novel mouse strains to enable doxycycline-inducible shRNA-mediated APC silencing specifically in the intestine and colon of adult mice. Most importantly, APC loss can be reversed simply by withdrawal of doxycycline, restoring the endogenous tumor suppressor network. In Aim 1 I will define the requirement for sustained APC loss in the maintenance of benign and dysplastic lesions, which represent the initiation of carcinogenesis in the colon and mirror the early stage tumors that develop in patients with Familial Adenomatous Polyposis (FAP). In addition, I will use the same transgenic mouse model to develop and characterize and surrogate ex vivo culture system to examine the molecular consequences of APC restoration in early stage disease. Aim 2 expands on the systems developed in Aim 1 to investigate how recurrent genetic alterations acquired during progression to invasive and metastatic CRC (oncogenic Kras mutation and loss of p53) affect the response to Wnt inhibition. As Kras and/or p53 mutations are observed in ~80% of APC mutant tumors, understanding how these events contribute to Wnt- dependency is critical for future clinical applications of Wnt-targeted therapy. Finally, through characterisation of the tumor response to APC restoration and detailed analysis of resistant or relapsed, Wnt-independent disease, I aim to identify candidate genes/pathways that mediate resistance to Wnt inhibition with the vision that they will represent attractive targeted for combinatorial therapeutic approaches. The work outlined in this proposal will provide key insights into the role of commonly mutated genes in CRC and define factors that are essential for the growth and survival of CRC cells, representing potential targets for rational drug development. The reagents, tools and knowledge developed through this work will form the foundation of a flexible and innovative research program that has the potential to contribute significantly to the understanding and treatment of many forms of cancer.